Novel Intelligent Pigging Tool For Deposit Inspection Using Electrical Tomography with High Computational Efficiency Enabled Through Neural Networks

2021 ◽  
Author(s):  
Antti Nissinen ◽  
Ossi Lehtikangas ◽  
Arto Voutilainen ◽  
Pasi Laakkonen ◽  
Anssi Lehikoinen
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Electrical Properties ◽  
Computational Efficiency ◽  
Deep Neural Network ◽  
Computation Time ◽  
Electrical Tomography ◽  
Deposit Thickness ◽  
The Neural Network ◽  
Pipe Line

Objectives/Scope Deposition inspection sensor based on electrical tomography has been proposed recently. In this work, a next generation electrical tomography sensor is introduced and a novel mathematical approach for the estimation of the deposit thickness is described. It is essential for the pipeline operators to keep the lines open for smooth flow and high flow efficiency. Deposit thickness, deposit type and location of deposit is required for optimal pipeline cleaning. The usage of chemicals as well as number of cleaning pig runs can be optimized based on the information that intelligent pig is giving. Methods, Procedures, Process In electrical tomography, electrodes are attached on the surface of the sensor and excitations are applied to some electrodes and responses are measured from other electrodes. The electrical properties of the medium are estimated based on these measurements. In pigging applications, the distribution of electrical properties between the PIG surface and metal pipe is estimated. The thickness and type of deposit (wax/scale) can be identified from the estimated electrical properties. In the proposed approach the estimation of the parameters is done by using a novel deep neural network based approach. In practice, number of measurements that are analyzed after each PIG run can be hundreds of thousands. The neural network based approach was chosen in order to achieve reasonable computational efficiency (computation time) in real applications with large amounts of data. Results, Observations, Conclusions The introduced sensor is for 12-inch lines and designed to be used when the oil line is filled with water. This sensor was tested in a laboratory test line with artificial deposit samples. After these tests and calibration, the sensor is deployed to be used in real pipe line inspections. The major challenges in pipe line runs include the movement of the sensor during measurements, electrical noise and changing excitations. In the neural network model, the position of the PIG is estimated simultaneously with the electrical properties and the effect of all these aforementioned uncertainties are also modelled. Based on the results conclusions can be drawn on the efficiency and performance using neural networks and the high suitability of electrical tomography for deposit mapping. Novel/Additive Information In this study, it is shown that intelligent pig based on the electrical tomography can be used reliable for deposit inspection. Furthermore, the computation approach based on the deep neural network is computationally efficient and it is tolerable for measurement noise and other uncertainties in real measurements.

scholarly journals Feature-Based Interpretation of the Deep Neural Network

Electronics ◽  
2021 ◽  
Vol 10 (21) ◽  
pp. 2687
Author(s):  
Eun-Hun Lee ◽  
Hyeoncheol Kim
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Neural Network ◽  
Lower Layer ◽  
Network Models ◽  
Neural Network Models ◽  
Network Pruning ◽  
The Neural Network ◽  
Single Instance ◽  
High Level

The significant advantage of deep neural networks is that the upper layer can capture the high-level features of data based on the information acquired from the lower layer by stacking layers deeply. Since it is challenging to interpret what knowledge the neural network has learned, various studies for explaining neural networks have emerged to overcome this problem. However, these studies generate the local explanation of a single instance rather than providing a generalized global interpretation of the neural network model itself. To overcome such drawbacks of the previous approaches, we propose the global interpretation method for the deep neural network through features of the model. We first analyzed the relationship between the input and hidden layers to represent the high-level features of the model, then interpreted the decision-making process of neural networks through high-level features. In addition, we applied network pruning techniques to make concise explanations and analyzed the effect of layer complexity on interpretability. We present experiments on the proposed approach using three different datasets and show that our approach could generate global explanations on deep neural network models with high accuracy and fidelity.

scholarly journals Scene Labeling using H-LSTM by Predicting the Pixels using Various Functions

2019 ◽  
Vol 8 (3) ◽  
pp. 1179-1185
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Graphical Models ◽  
Color Image ◽  
Computation Time ◽  
Reference Pattern ◽  
The Neural Network ◽  
Scene Labeling ◽  
Color Transformation ◽  
Color Enhancement

Scene Labeling plays an important role in Scene understanding in which the pixels are classified and grouped together to form a label of an image. For this concept, so many neural networks are applied and they produce fine results. Without any preprocessing methods, the system works very well compared to methods which are using preprocessing and some graphical models. Here the neural network used to extract the features is Hierarchical LSTM method, which already gives greater result in Scene parsing in the existing method. In order to reduce the computation time and increase the Pixel accuracy HLSTM is used with Makecform and Softmax functions were applied. The color transformation is applied using the Makecform function. The color enhancement of images has given object as input to H-LSTM function to identify the objects based on the referential shape and color. H-LSTM constructs the neural network by taking the reference pattern and the corresponding label as input. The pixels present in the neighbourhood identified with the help of neural network. In this method, the color image is converted into greyscale and then the Hierarchical LSTM method is applied. Therefore, this method gives greater results when it is implemented in Matlab tool, based on pixel accuracy and computation time when compared to other methods.

scholarly journals Book spine recognition with the use of deep neural networks

2020 ◽  
Vol 44 (6) ◽  
pp. 968-977
Author(s):  
M.O. Kalinina ◽  
P.L. Nikolaev
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Performance ◽  
Deep Neural Network ◽  
Deep Neural Networks ◽  
Original Data ◽  
Network Architectures ◽  
Data Set ◽  
The Neural Network ◽  
Processing Information

Nowadays deep neural networks play a significant part in various fields of human activity. Especially they benefit spheres dealing with large amounts of data and lengthy operations on obtaining and processing information from the visual environment. This article deals with the development of a convolutional neural network based on the YOLO architecture, intended for real-time book recognition. The creation of an original data set and the training of the deep neural network are described. The structure of the neural network obtained is presented and the most frequently used metrics for estimating the quality of the network performance are considered. A brief review of the existing types of neural network architectures is also made. YOLO architecture possesses a number of advantages that allow it to successfully compete with other models and make it the most suitable variant for creating an object detection network since it enables some of the common disadvantages of such networks to be significantly mitigated (such as recognition of similarly looking, same-color book coves or slanted books). The results obtained in the course of training the deep neural network allow us to use it as a basis for the development of the software for book spine recognition.

SCORING MODELING BASED ON NEURAL NETWORKS FOR DETERMINING A BANK BORROWER'S RATING

2020 ◽  
Vol 2020 (10) ◽  
pp. 54-62
Author(s):  
Oleksii VASYLIEV ◽  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Architecture ◽  
Statistical Data ◽  
Activation Function ◽  
Decision Making Process ◽  
Neural Network Architecture ◽  
Acceptable Accuracy ◽  
The Neural Network ◽  
Sigmoid Activation Function

The problem of applying neural networks to calculate ratings used in banking in the decision-making process on granting or not granting loans to borrowers is considered. The task is to determine the rating function of the borrower based on a set of statistical data on the effectiveness of loans provided by the bank. When constructing a regression model to calculate the rating function, it is necessary to know its general form. If so, the task is to calculate the parameters that are included in the expression for the rating function. In contrast to this approach, in the case of using neural networks, there is no need to specify the general form for the rating function. Instead, certain neural network architecture is chosen and parameters are calculated for it on the basis of statistical data. Importantly, the same neural network architecture can be used to process different sets of statistical data. The disadvantages of using neural networks include the need to calculate a large number of parameters. There is also no universal algorithm that would determine the optimal neural network architecture. As an example of the use of neural networks to determine the borrower's rating, a model system is considered, in which the borrower's rating is determined by a known non-analytical rating function. A neural network with two inner layers, which contain, respectively, three and two neurons and have a sigmoid activation function, is used for modeling. It is shown that the use of the neural network allows restoring the borrower's rating function with quite acceptable accuracy.

scholarly journals Discretization and machine learning approximation of BSDEs with a constraint on the Gains-process

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Idris Kharroubi ◽  
Thomas Lim ◽  
Xavier Warin
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Differential Equations ◽  
Numerical Experiments ◽  
Optimization Problem ◽  
Learning Approach ◽  
The Neural Network ◽  
Machine Learning Approach ◽  
Mesh Grid

AbstractWe study the approximation of backward stochastic differential equations (BSDEs for short) with a constraint on the gains process. We first discretize the constraint by applying a so-called facelift operator at times of a grid. We show that this discretely constrained BSDE converges to the continuously constrained one as the mesh grid converges to zero. We then focus on the approximation of the discretely constrained BSDE. For that we adopt a machine learning approach. We show that the facelift can be approximated by an optimization problem over a class of neural networks under constraints on the neural network and its derivative. We then derive an algorithm converging to the discretely constrained BSDE as the number of neurons goes to infinity. We end by numerical experiments.

scholarly journals Deep neural networks using a single neuron: folded-in-time architecture using feedback-modulated delay loops

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Florian Stelzer ◽  
André Röhm ◽  
Raul Vicente ◽  
Ingo Fischer ◽  
Serhiy Yanchuk
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Neural Network ◽  
Single Neuron ◽  
Deep Neural Networks ◽  
Back Propagation ◽  
Local Network ◽  
Multiple Time ◽  
Learning Tools ◽  
Back Propagation Algorithm

AbstractDeep neural networks are among the most widely applied machine learning tools showing outstanding performance in a broad range of tasks. We present a method for folding a deep neural network of arbitrary size into a single neuron with multiple time-delayed feedback loops. This single-neuron deep neural network comprises only a single nonlinearity and appropriately adjusted modulations of the feedback signals. The network states emerge in time as a temporal unfolding of the neuron’s dynamics. By adjusting the feedback-modulation within the loops, we adapt the network’s connection weights. These connection weights are determined via a back-propagation algorithm, where both the delay-induced and local network connections must be taken into account. Our approach can fully represent standard Deep Neural Networks (DNN), encompasses sparse DNNs, and extends the DNN concept toward dynamical systems implementations. The new method, which we call Folded-in-time DNN (Fit-DNN), exhibits promising performance in a set of benchmark tasks.

Application of neural networks in predictions of brake wear particulate matter emission

2021 ◽  
pp. 095440702110363
Author(s):  
Saša Vasiljević ◽  
Jasna Glišović ◽  
Nadica Stojanović ◽  
Ivan Grujić
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Particulate Matter ◽  
Traffic Control ◽  
World Health ◽  
Vehicle Speed ◽  
System Pressure ◽  
The Neural Network ◽  
Vehicle Systems ◽  
Brake Wear

According to the World Health Organization, air pollution with PM10 and PM2.5 (PM-particulate matter) is a significant problem that can have serious consequences for human health. Vehicles, as one of the main sources of PM10 and PM2.5 emissions, pollute the air and the environment both by creating particles by burning fuel in the engine, and by wearing of various elements in some vehicle systems. In this paper, the authors conducted the prediction of the formation of PM10 and PM2.5 particles generated by the wear of the braking system using a neural network (Artificial Neural Networks (ANN)). In this case, the neural network model was created based on the generated particles that were measured experimentally, while the validity of the created neural network was checked by means of a comparative analysis of the experimentally measured amount of particles and the prediction results. The experimental results were obtained by testing on an inertial braking dynamometer, where braking was performed in several modes, that is under different braking parameters (simulated vehicle speed, brake system pressure, temperature, braking time, braking torque). During braking, the concentration of PM10 and PM2.5 particles was measured simultaneously. The total of 196 measurements were performed and these data were used for training, validation, and verification of the neural network. When it comes to simulation, a comparison of two types of neural networks was performed with one output and with two outputs. For each type, network training was conducted using three different algorithms of backpropagation methods. For each neural network, a comparison of the obtained experimental and simulation results was performed. More accurate prediction results were obtained by the single-output neural network for both particulate sizes, while the smallest error was found in the case of a trained neural network using the Levenberg-Marquardt backward propagation algorithm. The aim of creating such a prediction model is to prove that by using neural networks it is possible to predict the emission of particles generated by brake wear, which can be further used for modern traffic systems such as traffic control. In addition, this wear algorithm could be applied on other vehicle systems, such as a clutch or tires.

scholarly journals Deep neural network analysis of nanoparticle ordering to identify defects in layered carbon materials

2021 ◽  
Author(s):  
Daniil A. Boiko ◽  
Evgeniy O. Pentsak ◽  
Vera A. Cherepanova ◽  
Evgeniy G. Gordeev ◽  
Valentine P. Ananikov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Network Analysis ◽  
Carbon Material ◽  
Deep Neural Network ◽  
Carbon Materials ◽  
Material Surface ◽  
Neural Network Analysis ◽  
Pd Nanoparticle ◽  
Sem Imaging

Defectiveness of carbon material surface is a key issue for many applications. Pd-nanoparticle SEM imaging was used to highlight “hidden” defects and analyzed by neural networks to solve order/disorder classification and defect segmentation tasks.

Estimating Traffic Volume on Minor Roads at Rural Stop-Controlled Intersections using Deep Learning

2019 ◽  
Vol 2673 (4) ◽  
pp. 108-116 ◽  
Author(s):  
Mostafa H. Tawfeek ◽  
Karim El-Basyouny
Keyword(s):  
Neural Network ◽  
Linear Regression ◽  
Network Model ◽  
Neural Network Model ◽  
Deep Neural Network ◽  
Safety Management ◽  
Estimation Method ◽  
Estimation Model ◽  
The Neural Network ◽  
Annual Average Daily Traffic

Safety Performance Functions (SPFs) are regression models used to predict the expected number of collisions as a function of various traffic and geometric characteristics. One of the integral components in developing SPFs is the availability of accurate exposure factors, that is, annual average daily traffic (AADT). However, AADTs are not often available for minor roads at rural intersections. This study aims to develop a robust AADT estimation model using a deep neural network. A total of 1,350 rural four-legged, stop-controlled intersections from the Province of Alberta, Canada, were used to train the neural network. The results of the deep neural network model were compared with the traditional estimation method, which uses linear regression. The results indicated that the deep neural network model improved the estimation of minor roads’ AADT by 35% when compared with the traditional method. Furthermore, SPFs developed using linear regression resulted in models with statistically insignificant AADTs on minor roads. Conversely, the SPF developed using the neural network provided a better fit to the data with both AADTs on minor and major roads being statistically significant variables. The findings indicated that the proposed model could enhance the predictive power of the SPF and therefore improve the decision-making process since SPFs are used in all parts of the safety management process.

scholarly journals A Neural Network Decomposition Algorithm for Mapping on Crossbar-Based Computing Systems

Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1526 ◽  
Author(s):  
Choongmin Kim ◽  
Jacob A. Abraham ◽  
Woochul Kang ◽  
Jaeyong Chung
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Predictive Performance ◽  
Fixed Number ◽  
Decomposition Algorithm ◽  
Computing Systems ◽  
Network Decomposition ◽  
Von Neumann ◽  
Analog Computing ◽  
The Neural Network

Crossbar-based neuromorphic computing to accelerate neural networks is a popular alternative to conventional von Neumann computing systems. It is also referred as processing-in-memory and in-situ analog computing. The crossbars have a fixed number of synapses per neuron and it is necessary to decompose neurons to map networks onto the crossbars. This paper proposes the k-spare decomposition algorithm that can trade off the predictive performance against the neuron usage during the mapping. The proposed algorithm performs a two-level hierarchical decomposition. In the first global decomposition, it decomposes the neural network such that each crossbar has k spare neurons. These neurons are used to improve the accuracy of the partially mapped network in the subsequent local decomposition. Our experimental results using modern convolutional neural networks show that the proposed method can improve the accuracy substantially within about 10% extra neurons.

Sign in / Sign up

Export Citation Format

Share Document